Akai GX-636 Repair

[Ray122]

Hello all! Been working on an Akai GX-636 lately and thought I'd post my findings to maybe help anyone who's facing similar issues.

The original issue was that the unit would only play for a few minutes before it would disengage playback. It took me a while to troubleshoot this one, and I found that the long I worked on it, the shorter period of time I would have before the unit would disengage from play mode. After even more time, the capstan stopped spinning altogether.
This turned out to be a FR1 (fused resistor) and R31. Evidently they were both slowly increasing in resistance until the fused resistor finally popped.
R31 is 56K as noted on the schematic, FR1 (which is labeled as R30 on the PCB) is 100 ohms. I mention this because on the schematic FR1 is noted as 100(K). I very wrongly assumed this meant 100K, but it is in fact 100 ohms, and it turns out the (k) refers to the wattage. Although I don't believe this is explicitly noted on the schematic, unless I'm blind to it after staring at it for so long.

It also took me a long time to figure out how the whole capstan motor control circuit works. If you look at the schematic, the motor windings are in series with a bridge rectifier before returning to the AC neutral of the transformer. Interestingly, the DC legs are connected to the emitter and collector of TR901. Essential TR901 is used to 'short out' for lack of a better term the DC side of the rectifier, which draws the specific current through the AC side of the rectifier to spin the capstan motor at the precise speed it needs to be at.

I'm currently working on sorting out why the audio muting circuit wants to stay active all the time. In play mode, pin 38, which connects the output of the digital logic circuit to the audio muting circuit is low at 1.7V and high at 5V when in fast forward/rewind or stop. Evidently the 1.7V is still too high for the audio muting circuit. TR19 of that circuit is looking for something below .5V in order to 'open' and allow audio to pass freely. My work around is to use two diodes to get a 1.4V drop off of pin 38 so that the signal going to TR19 is low enough to function normally. But if anyone smarter than me can make any recommendations as to why I might be seeing too high of a 'low' voltage there, I'd appreciate it!

I'll keep you posted on this one.

 

[Ray122]

Okay, took another look at it this afternoon. I started poking around IC 5, looking at the voltages inputing from pin 11 and outputting from pin 10 for the play back mute function. I was noticing that even the input of pin 11 would swing between 5V (high) and 1.7V (low), when according to the schematic a low input should read .7V. Unable to think of anything else that might cause this issue, I decide to check the VCC and GND pins on IC 5, thinking maybe if the VCC input voltage is too high for some reason, that could be causing the outputs to output a higher voltage than normal. This took a little investigative work because the schematic doesn't explicitly notate the VCC input (Pin 14) and the GND for IC 5 (pin 7). I had to look at the PCB layout to determine where these pins where connected and if they had a direct connection to GND and the 5V output from power supply like you'd assume they would. In doing this, I discovered my initial hunch that the VCC input was too high was incorrect, I did discover the pin 7 for GND didn't have continuity to the chassis. Exploring the PCB a little further turns out one of the screw points that secures the PCB to the chassis has a metal plate for making the ground contact for the circuit... and since I had the whole board pulled out of the chassis so I could probe around to solve my original issue, I wasn't grounding the circuit correctly... dumb oversight, but hopefully maybe some insight that might help someone else out some day.

Needless to say after grounding the circuit correctly, the low is now at .7 volts like it should be, and the muting circuit works great without that crazy double diode work around I was contemplating implementing.

Now, when I put it into record mode, I only get the right channel going to tape. Tape and Source Monitor work just fine, so it's not obvious (yet) where I'm losing the left channel. So one more gremlin to chase, but hopefully it's something simple like a dirty switch or something... Stay tuned.

 

[Ray122]

Tonight I checked the switches assuming I lost contact for the left channel somewhere through one of the multiple mechanical switch contacts it passes through, but as I rocked the switches around manually no sign of the left channel coming back. The other odd thing I noticed was that I was able to record both left and right channels to tape without an issue if I ran the unit in reverse. (with the take up spoon on the left) Since the unit head two sets of play, record and erase heads, one for set for forwards play and one set for reverse play, it seemed to me the issue had to be with the tape head itself since jogging the mechanical switch did nothing, and all the audio flowed just fine through all the amplifier circuit when recording in reverse. I decided to just examine the record and play heads to make sure a wire hadn't broken off the heads or something, and just by pressing on the solder pad of the record head of the left channel magically came back. The solder joint looks fine... so not totally sure what the deal is there, but seems to work great now! I'm going to keep an eye on it over the next few days and if it acts up again I'll reflow the joint and see if that helps.

Anyone have any issues with record or play heads going bad like that?

This will probably be the last post I make on this thread for this project since everything is working now, but if anyone stumbles upon this thread while tackling their own 636 and needs some help, feel free to reach out!